Weekly Update 10

So it’ been awhile. Almost a month.

My last post had a treat enclosed, this one however has a more of a sour note for those who follow this blog. Although that isn’t many, everyone counts. I have decided to cut down on the amount of posts I make as you can already tell. It was a good experiment to do weekly updates (3 months was a good run) but I either became a slave to them and I was going to be reaching that point of quantity over quality. So from here on in, no more weekly updates. Just updates when it feels right with the right amount of content. At the very least Iwill try to keep it at one post a month but even I feel that might be a bit low so we will see how it goes from here on in, no more set deadlines for sure.

Final (Low Rez) Renders

 

The difference with this beard from the old one is it is control via custom sculpted guide curves (around 500). Not only is it easier to sculpt but also when it dealing with to penetrations. The curves were sculpted in Blender.

Render times kept to a minimum with the shadow volume technique that i described in my tutorial. The body hair, is pure PBR with no volume based shadows.

finalbody

These pics are low rez but i am currently rendering HD turntables and camera moves, posing. Also i have been uploading all work-in-progress pictures of this character to my Google account so if you would like to view those images in the future you can do so from one place soon. It includes all images featured in the blog along with some others that have not been published.

 

Tutorials Response with Q/A

After the tutorial i posted i was very surprised at the response. I hope everyone who watched it gained something from it. I have decided to respond to the questions in this post, continue if you would like to see the questions and answers.

Continue reading “Weekly Update 10”

Weekly Update 10

Weekly Update 09

So it’s a little late i admit, but i have a treat for you.

So the past week has been interesting to say the least. Many XSI/SI users are looking for a new home. It seems Houdini looks like a viable choice for many and not only are the odforce and SESI forums more active (although they already were to be honest), there has been a vocal response from a few SESI employees stating they want to help not just by setting up a forum but also in the package. Encouraging discussion on how to improve Houdini.

Not only that, but Houdini users are also welcoming this new crowd and doing their best to help them  with the transition. I have even decided to join in that effort.

A user on od|force was asking how would one go about connecting guide curves into the Fur module in Houdini. I responded with this video.

Continue to see a the breakdown of chapters.

Continue reading “Weekly Update 09”

Weekly Update 09

Weekly Update 02

Still here? Here is a rundown of the week’s tinkering.

Fur

So bar any tweaks, texturing is pretty much done but I think I will be messing with the shader forever to be honest. I’m still not happy with the material.

The character of Virgil was intended to be part of a tribute to Dante’s Inferno but my intentions shifted a while ago to make him more of a homeless person. So putting him in a setting of the harsh cold streets, I think it would be remiss not to give him a beard of some sort. Perfect opportunity to utilize Houdini’s Fur module, wasn’t sure if i wanted a big log beard or a more conservative style. What do you think. All of the beard and body hair was groomed in Houdini.

Continue reading “Weekly Update 02”

Weekly Update 02

Bee Project: Part Two – Modelling

Hello there,

So here comes the obligatory apology. Sorry. I’ve been away for a while over the summer.

But you guys better be grateful for this post. Ive spent allot of time gathering up the media that went into this post so i hope you enjoy it all.

In part one I started by talking about how people got to this site, so with this post i have to say the thing on this blog that gets the most hits has to be the Fur network setup pic.

Since then, Houdini has matured into version 9.5, so the next part of this series will no doubt have some different methodologies. I have played a tiny bit with fur in the current version and it is different. More on that next post

Final mode


Lets begin with some stats then. It was modeled entirely in Wings3d. It currently weighs in at around 20k polys, there arent any UV’s. This might be covered in the next post (im not sure because UV’s are a pretty boring process, i dont know if anyone cares to read about it).

Im gald i got to finish this thing but there is a lot of research behind it so sit back and enjoy.

Notes & Research

Before i started i knew this was going to be difficult just for the fact that its hard to see whats going on under all that fur. So i gather a few diagrams and basic reference images and drew what i felt was a good shape for the body, wings, mouth etc.

I didnt do this so i could provide myself with image planes or something definitive to work from. This was basically to understand as much as i could and also point out ideas and issues long in advance.

These drawings/sketches was a good push off point to know what references i needed more of and other research that would be useful later in the project.

ie biological Iridescence for the wings.

If you are going to do this, always remember these images are for you and not for anyone else. These dont have to be perfect or presented in a nice way. It is just an outlet for getting your ideas and thoughts down on paper.

Here my pages from this project.

Reference

So from the notes and sketches i created i knew what i needed. One thing the interest provides is a wealth of choices. I probably have 200 images of bee’s in my reference folders and came across 1000 images in total via google image search. The main difficulty is continuity. There are so many types of bee’s in the world, so when you enter “honey bee”, you will come across 50 variations. Even if they are the same species, there is a lot of differences.

So the approach i decided to take was to gather groups of photos. If i found an image of a bee, i would try to find out if they are part of a collection.

The best place is to find these types of collections is on forums. There are forums for just about anything and the best place i found were photography and specialist bee forums. But When looking for stuff on forums via google always put “showthread” with whatever you need to find in the search box. This will always point to a thread on a forum and if its a good lead, then you can browse that site.

Then when I had found as many collections as I could take then I would gather specific body parts. Body, Abdomen, mouth, wings, eyes.

Here is a sample of the references I used.

This was in fact was the easy part. But the overall shape of the bee was still a bit of enigma since its covered by so much “fur”. But i have found that the hardest object/people to model isn’t because they are hard surfaces to mimic, its just you don’t have enough reference. And when i did gather the reference for this project i made sure everything was organized so i knew where to get certain images from.

One thing that threw me off was what type of bee am i creating. Was it a drone, a worker or the queen?

See if you can notice the differences between them (especially between the drone and worker).

makin’ of

Before I talk about the method of modelling this new found monster, i have to say get into the habit of making incremental saves. This project alone. Which was modelled in entirely in wings, has just under 100 saves. Not only is it easier to find the most current file, especially when using other applications but also if the original file becomes corrupted, you always have a place to fall back on.

It would be very awkward for me to describe how i model the actual bee, without getting very lengthly. So i have provided a video of the various files with a look at the final model.

Enjoy!
Vodpod videos no longer available.
Vimeo Link for Google Reader Users

Vimeo, is crap, i know. I am trying to make it host it on veoh like the Angry Boy video, but their servers are slow or something. But you can download the file now (when you host with Vimeo).

What’s Next

The next part of this saga will probably be the last part for a while.

It will be the creation and experimentation process of applying fur to the model. I will explain it in as much detail as possible. The good parts, the bad parts.

And basic rendering, because in part 3 will cover the rendering and shader process. I have not yet decided to write the shaders in VEX or make it in VOP’s.

I will try to make some sort of iridescence shader for the wings, a procedural shader for the body and a geometry/displacement shader for the compound eye.

In part 4, i will again try to cover the rigging process of the honey bee. I chose this project for a particular reason, it incorporates many areas of CG whilst remaining relatively simple. The rigging process should be pretty simple but i have an idea for multiple rigs.

In part 5 I will animate it in a scene. The scene will contain a two characters, a Bee and a person and an enviroment. This will take quite a while.

So by this time next year part 5 should be coming out. 🙂

… hopefully hehe

by now you should be satisfied with this post.

I will say that this post was delayed a little more by this weeks events. What am i talking about…

Siggraph 2008

So every year, one week in August is an opportunity for almost every major and minor cg software company to show whats coming up in the future months. Its like E3 or GDC. More like GDC because people also come to show off their technology and how they did it.

Autodesk’s Users Group will probably be the talking point along with Zbrush (which at the time of writing this has yet to publicly discuss Zbrush 3.5 ). But autodesk has made a major effort for those who cannot attend to put the entire week long event online (live streaming) and the User’s Group Presentation was about 4 hours long (I watched 90%, just before the sun came up 🙂 ).

There were no crashes unfortunately hehe, but there was on Tuesday on the live stream. I just caught a guy showing of the rendering capabilities (which were amazing) and the program died. All he kept saying was “this is bad, this is bad”.

Hehe poor fella

I gotta say though I agree with many that the next version of Mudbox looks like it could have the edge on Zbrush, but i cant rule anything out until Pixologic say anything.

Phew!.

Hope you enjoyed the read, images and videos. The next post might be a comp of speed sculpts or a shader i want to write/create. These project post take a lot of time so im sure you will appreciate if i have to put in one or two posts about other stuff im working on.

Until then.

Take care.

-andy

Bee Project: Part Two – Modelling

Bee Project: Part One – Conception & Preparation

Hey guys,

Just a quick preface:

Its funny how many people are reaching this site via search engines with the terms “fur 3d” or “render fur tutor Renderman”. So i think its pretty obvious that most of the people reaching this site are looking to read some stuff on Renderman or Houdini.

When looking at the stats of yesterday and today please refrain from laughing at the number of views. hehe 🙂 (most of the hits i get on the site are mostly referred links rather than via a search….. but its still pathetic :))

So onto the real topic here, I hate to disappoint you guys for another post, but i want to document this thing properly. (I want to be proud of the way I post this material).

Originally i was going to write 3 posts on making of the Honey Bee. But I want this to be accessible to everyone and also show as much as i can about my process.I cant really say how many posts this will take but this first entry will revolve around the conception/preparation and research but also my motives, reasons and some early choices. This is the introduction to the Honey Bee project but it also shows the same setup I would use for any other project.

Origin

Ok, so you know I am a character person. There are few (technical) elements that make up a character in my world, polygons (shape & form), shaders (skin & misc object) and Fur (hair or just plain fur). These are the things I want to master. You can understand now, why i want to learn shader writing (and lighting). Eventually, should I master them (doubtful), or the only thing left for me to do before I become a master is to practice practice practice, then I want to move into rigging and animation. Yeah its a lot of stuff but i love Cg, not just modelling, EVERYTHING. I also understand that i need to master one thing first before moving on and the chances are, no matter how good i get in the other areas i wont be as good as i am in modelling. But its still possible to become a Jack of all trades and master of 2 or 3 (the restriction doesnt have to be narrowed to just 1).

So for a while now, Fur has been the one quest i have yet to have ANY success in. Houdini (which is my primary app) has had Fur since 9.0 but it still hasn’t quite got the control i want (another reason to learn VEX is all the control shaders are written in CVEX).

But recently on the odForce forum’s Mark Visser created Fuzzy Toolkit (btw he was the principal architect of the custom software used to groom and render all of those wonderful furry and feathered critters on C.O.R.E’s movie, “The Wild”). He really understands both the artists needs and pipeline requirements.

The thing is there is a problem with the software / my setup, he is trying to find out which one it is.

But my hunger for Fur wont go away.
Like I said earlier Houdini does offer a solution, albeit limited in the grooming department but it is the only option right now.

To be honest though, if it wasn’t for a hot sunny Sunday I probably wouldn’t be doing this project. I was watching Tv and the biggest bee flew in through a window and I freaked out. I closed all the door and grabbed a cloth and whipped it (just like Indy). It fell to the ground (almost with a thud, it was that big).

I did feel a little bad but I’ve always wanted a bee to study so that one day I could model it, because in the past when I’ve tried to model such creatures and I cant find enough reference of the same bee or insect.
So I put it on a sheet of paper and put a glass on top. Took about a day and a half to die (which was a little bit cruel) but now I have a bee to examine. 🙂

So i went into project mode. I thought of certain technical/aestetic elements of this bee.

Here they are.

  • Predominately fur covered (at least 70%)
  • Interesting/appealing creature/character (doesn’t mean it has to be liked, just something that sits well with the viewer)
  • Challenging modelling-wise
  • Wouldn’t be to hard to rig (possible first rigging project )
  • The possibility to create some new and interesting shaders
  • In terms of fur grooming, it would have to be pretty basic

Even though the fur grooming will be pretty limited, I can still learn a lot in terms of Fur rendering.

Perfect.

So really I started this thing backwards, but when I intentionally create something, i have a list like the one above of what characteristics the subject should possess.

The chosen creature is, a Honey Bee! (like you didn’t know judging from the title of this post).

Since I am taking this project seriously, I bought a Notepad so could jot down any ideas, sketches and well…. notes. Without sounding geeky (I writing a blog, how geeky can I get :)), this is an awesome Notepad.

ok I think I’m gonna call it there. Time of post 11:42pm

Hope this helps.

Next post, Modelling.

Should be out in the next few days.

Really! 🙂

-andrew

Bee Project: Part One – Conception & Preparation

1000+ Hits and Fur Virgin, no more!

This post is lengthly so grab a cup of something.

Yes thats right, this silly little blog that still has a wordpress affix has had over 1000 hits. I guess its a bit of an achievement (but who give a crap, most sites gets that in a day). Even though less than 1% has commented, should i take that as a hint that the rest of the web hits are just people who stumble here by accident and leaving swiftly? Let me know 🙂
But what the the point of this entry, well i have had no reason to update except ive been working on a particular project…
You may remember the post a while ago about me getting frustrated with Houdini Fur. Well since then SideFX has released a multitude of written tutorials on the mechanics of the Fur Sop and Fur Procedural (although if SideFX is listening i think the people would love to see a workflow walkthrough).
So here are my pathetic results after nearly 30 day of tinkering.

The model was made in Wings 3d, and took a night to make (2 hours tops). Someone please let me know where to find good mouse reference. I spent nearly 2 hours alone looking for decent images, and i got a handful and the rest were of the mechanical variety. So this is what i have to so for my efforts, not much but i think ive made some ground especially with the last image (below).

So to help any Houdini wannabe’s outthere, like myself. Let me explain how i got from the model to the last image. ughhhhh… here is the network (much easier).

So for those who use houdini you shouldnt be too scared, the nulls are referenced by an object merge in another object, but to be totally honest you should check out the written tutorials that come in the help files. go through them, dont just look at the files go through them (it really helps).
I guess the only difference from the example files and my setup is the Fur Procedural has varying root width. I have painted the model with a Paint SOP and that is fed into a Attribute Create SOP which uses the painted points to control the density, (although im thinking of using a texture map to have greater control, but this method is much faster). Im going to try and use this method to control the length but the model obviously isnt that high rez so it might be a bit hard. BTW the parameters needed to control these options are in the fur sop help files. When the skin geoemtry is reference its carried over to the Fur procedural (to check MMB over the Skin GEO to see if it has eg FURDENSITY, FURLENGTH). You need to also reference them in the Fur SOP for preview purposes.
The normals have been combed with a Comb SOP to govern the direction of the fur. If you have a symmetrical model be sure to use the mirror tab (also used it in the Paint SOP).In a few tests i did edit the copied lines, just before the Fur sop, but its was way too labor intensive. There needs to be some changes made to the Edit SOP by the developers which allows the Brush option to have a move/tweak option, kinda like Zbrush’s Move brush. Its more like sculpting at the moment.

In a nut shell, thats it, and im still working on the lighting, dynamics and rigging, so its an expansive project but one that is really getting me around Houdini.
So there should be enough in there for users who have gone through the tutorials but need some sort of insight. One piece of ….. Two pieces of advice i would give when making Fur is, Dont give up and dont be afraid to start all over again (you learn something new each time).

After all that i have to say, making an almost completel Fur covered model has to be one of the most difficult things to do in CG. Even big studios have trouble with it. I know this isnt a news blog in particular but one fella who knows Fur is Andy Boyd (who was recently interviewed by odForce). Here is a few pics he did for Framstore in the rexona commercial.

Mandrill

Andy has since moved to LA where he now works at Method Studios. And here is one of the first things he has worked on there.

Method Studios

Interestingly, here is Framestore again, without Andy on the Super Squirrels commercial.

Framestore

Now im not trying to stir some shit up here. You can see what im saying without me having to say it. But the factors in what might seem a simple observation are pretty influential. You have to consider the people on the team, the software/pipeline, the client and of course the biggest factor of all TIME. Now im sure Andy wont take all the credit, not alot of people would (If you accept credit when something good happens, then you gotta accept the blame when something bad happens,) but I think we know Andy had a big part to play.
One thing i will say is the Fur dynamics in the Super Squirrels spot was amazing, and really made the sequence hold up. Maybe thts why as a still it might not look overly impressive.
Framestore has created some of the best work in Europe, i just hope that Andy taught his techniques to few people before he left (and that they were tied up on some other project so they couldnt work on Super Squirrels spot).

Again I dont know the full story, its just my take on it.

So thats it for what must be the longest post on the site. I will no doubt see you within another 30 days.

Just joking, see ya soon

-andy

1000+ Hits and Fur Virgin, no more!